Exploring the use of underwater video for fish monitoring in the Kawarthas: part 2
- grahamdraby
- 1 day ago
- 3 min read
By Jacob Bowman
Jacob was an undergraduate student in the Raby Lab when he conducted this work, and is currently an M.Sc. student working out of the Harkness Laboratory of Fisheries Research
It has been 3 years since my first Stoney Lake blog post about using remote underwater video (RUV) to study fish populations in the Kawarthas. Since we have recently published a scientific paper resulting from this work, Graham and I thought an updated blog post would be fitting.
Fish have tremendous ecological and cultural value in the Kawarthas, so monitoring their populations is important. Conventionally, we assess the status of fish populations using methods like gill netting, that involve catching lots of fish, leading to injuries and deaths. RUV could be used to monitor populations without the need to catch any fish. There are two main things that fish surveys aim to measure: species richness (how many species are present) and abundance (the number of individuals of each species). The objective of my project was to assess how the estimates of species richness and abundance from RUV compare to those from other methods. We sampled Upper Stoney Lake and the Otonabee River to compare RUV to three more conventional techniques for assessing fisheries: minnow trapping, snorkel transects, and seine netting (netting at Otonabee sites only). Minnow trapping and seine netting involving capturing fish, whereas snorkelling is a non-invasive method like RUV. At each site, we would deploy a RUV system (Figure 1) for 1-hour, recording continuously. We then sampled the same areas with the other methods to get estimates of species richness and abundance.
We found that RUV estimated similar or higher species richness when compared with minnow traps and snorkeling, meaning that overall it was as good or better than these other methods. However, seine netting caught more species than RUV. Small-bodied species like minnows were detected poorly by RUV because they were difficult to identify in the camera field of view, relative to when they could be examined when captured. The seine net we used covered a large area (100 m2) and had a fine mesh, so fish in that area were unlikely to escape or be missed.
Because fish can frequently swim through the camera field of view, it is difficult to count individuals (there are some sample videos below). The simplest way to get an estimate of abundance with RUV is with MaxN, which is based on the maximum number of individuals of a given species in the field of view at any one time during the recording. We found that maxN was correlated with estimates of abundance from seine netting for only 4 out of 11 species, suggesting that maxN is only a useful indicator of abundance for some species.
Another finding was that RUV was useful for detecting a dramatic population increase of the invasive round goby (Figure 2). In 2022 when we began the RUV study, round goby were detected for the first time north of Peterborough, in the Otonabee River. When invasive species enter new systems, they typically exhibit a population ‘boom’, where they become hyper abundant. Between 2022 and 2023, a boom did occur and was obvious with both RUV and seine netting. This suggests that RUV may be useful for detecting major changes in the abundance of fish, although this is an extreme example of population change.
We concluded that RUV is a promising tool for monitoring fisheries because of its versatility and ease of use. RUV may be especially useful for surveying sensitive and at-risk species, in combination with other techniques, or for community groups. Thanks to members of the Stoney Lake community, namely Bill and Gail Szego and Ralph and Carol Ingleton for supporting this project through donations to the Trent Aquatic Research Program.
Some video samples from the project:
Comments